{"url":"/task/feature-upsampling","name":"Feature Upsampling","slug":"feature-upsampling","description_markdown":"Deep features are a cornerstone of computer vision research, capturing image semantics and enabling the community to solve downstream tasks even in the zero- or few-shot regime. However, these features often lack the spatial resolution to directly perform dense prediction tasks like segmentation and depth prediction because models aggressively pool information over large areas. Feature Upsampling aims to recover this missing spatial resolution without impacting the space of the original deep features.","categories":[{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Methodology","url":"/area/methodology"},{"name":"Natural Language Processing","url":"/area/natural-language-processing"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":25,"papers_with_code":22,"benchmarks":1,"benchmark_tables_in_archive":1,"benchmark_tables_shown":1,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":2,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/feature-upsampling-on-imagenet","slug":"feature-upsampling-on-imagenet","dataset":"ImageNet","dataset_url":"/dataset/imagenet","rows_in_archive":8,"metrics":["ADCC","Average Drop","Average Increase"],"first_row_in_archive_order":{"model":"JAFAR","paper_title":"JAFAR: Jack up Any Feature at Any Resolution","paper_url":"/paper/jafar-jack-up-any-feature-at-any-resolution-1","paper_date":"2025-06-10","arxiv_id":"2506.11136","code_links":[{"title":"PaulCouairon/JAFAR","url":"https://github.com/PaulCouairon/JAFAR"}],"syntology":null}}],"datasets":[{"url":"/dataset/imagenet","name":"ImageNet","full_name":"","num_papers_in_archive":15430},{"url":"/dataset/gensc-6g","name":"GenSC-6G","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[{"url":"/task/representation-learning","name":"Representation Learning"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":22,"of":22,"tagged_in_all":25,"items":[{"url":"/paper/deep-image-prior","title":"Deep Image Prior","date":"2017-11-29","arxiv_id":"1711.10925","repositories_listed":14,"syntology":{"n":6,"n_ran":5,"n_unverified":1,"n_pointer_only":6}},{"url":"/paper/carafe-content-aware-reassembly-of-features","title":"CARAFE: Content-Aware ReAssembly of FEatures","date":"2019-05-06","arxiv_id":"1905.02188","repositories_listed":3,"syntology":null},{"url":"/paper/loftup-learning-a-coordinate-based-feature","title":"LoftUp: Learning a Coordinate-Based Feature Upsampler for Vision Foundation Models","date":"2025-04-18","arxiv_id":"2504.14032","repositories_listed":2,"syntology":{"n":6,"n_ran":2,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/featup-a-model-agnostic-framework-for","title":"FeatUp: A Model-Agnostic Framework for Features at Any Resolution","date":"2024-03-15","arxiv_id":"2403.10516","repositories_listed":2,"syntology":{"n":5,"n_ran":4,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/on-point-affiliation-in-feature-upsampling","title":"On Point Affiliation in Feature Upsampling","date":"2023-07-17","arxiv_id":"2307.08198","repositories_listed":2,"syntology":{"n":8,"n_ran":5,"n_unverified":3,"n_pointer_only":5}},{"url":"/paper/sapa-similarity-aware-point-affiliation-for","title":"SAPA: Similarity-Aware Point Affiliation for Feature Upsampling","date":"2022-09-26","arxiv_id":"2209.12866","repositories_listed":2,"syntology":null},{"url":"/paper/jafar-jack-up-any-feature-at-any-resolution-1","title":"JAFAR: Jack up Any Feature at Any Resolution","date":"2025-06-10","arxiv_id":"2506.11136","repositories_listed":1,"syntology":null},{"url":"/paper/benchmarking-feature-upsampling-methods-for","title":"Benchmarking Feature Upsampling Methods for Vision Foundation Models using Interactive Segmentation","date":"2025-05-04","arxiv_id":"2505.02075","repositories_listed":1,"syntology":null},{"url":"/paper/lda-aqu-adaptive-query-guided-upsampling-via","title":"LDA-AQU: Adaptive Query-guided Upsampling via Local Deformable Attention","date":"2024-11-29","arxiv_id":"2411.19585","repositories_listed":1,"syntology":null},{"url":"/paper/lighten-carafe-dynamic-lightweight-upsampling","title":"Lighten CARAFE: Dynamic Lightweight Upsampling with Guided Reassemble Kernels","date":"2024-10-29","arxiv_id":"2410.22139","repositories_listed":1,"syntology":null},{"url":"/paper/efficientcd-a-new-strategy-for-change","title":"EfficientCD: A New Strategy For Change Detection Based With Bi-temporal Layers Exchanged","date":"2024-07-22","arxiv_id":"2407.15999","repositories_listed":1,"syntology":null},{"url":"/paper/fade-a-task-agnostic-upsampling-operator-for","title":"FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures","date":"2024-07-18","arxiv_id":"2407.13500","repositories_listed":1,"syntology":null},{"url":"/paper/a-refreshed-similarity-based-upsampler-for","title":"A Refreshed Similarity-based Upsampler for Direct High-Ratio Feature Upsampling","date":"2024-07-02","arxiv_id":"2407.02283","repositories_listed":1,"syntology":{"n":8,"n_ran":6,"n_unverified":2,"n_pointer_only":8}},{"url":"/paper/lift-a-surprisingly-simple-lightweight","title":"LiFT: A Surprisingly Simple Lightweight Feature Transform for Dense ViT Descriptors","date":"2024-03-21","arxiv_id":"2403.14625","repositories_listed":1,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/arcnet-an-asymmetric-residual-wavelet-column","title":"ASCNet: Asymmetric Sampling Correction Network for Infrared Image Destriping","date":"2024-01-28","arxiv_id":"2401.15578","repositories_listed":1,"syntology":null},{"url":"/paper/learning-to-upsample-by-learning-to-sample","title":"Learning to Upsample by Learning to Sample","date":"2023-08-29","arxiv_id":"2308.15085","repositories_listed":1,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/seaformer-squeeze-enhanced-axial-transformer","title":"SeaFormer++: Squeeze-enhanced Axial Transformer for Mobile Visual Recognition","date":"2023-01-30","arxiv_id":"2301.13156","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/fade-fusing-the-assets-of-decoder-and-encoder","title":"FADE: Fusing the Assets of Decoder and Encoder for Task-Agnostic Upsampling","date":"2022-07-21","arxiv_id":"2207.10392","repositories_listed":1,"syntology":{"n":3,"n_ran":0,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/local-and-global-gans-with-semantic-aware","title":"Local and Global GANs with Semantic-Aware Upsampling for Image Generation","date":"2022-02-28","arxiv_id":"2203.00047","repositories_listed":1,"syntology":null},{"url":"/paper/deep-vit-features-as-dense-visual-descriptors","title":"Deep ViT Features as Dense Visual Descriptors","date":"2021-12-10","arxiv_id":"2112.05814","repositories_listed":1,"syntology":{"n":6,"n_ran":1,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/joint-denoising-and-demosaicking-with-green","title":"Joint Denoising and Demosaicking with Green Channel Prior for Real-world Burst Images","date":"2021-01-25","arxiv_id":"2101.09870","repositories_listed":1,"syntology":null},{"url":"/paper/a-unified-multi-scale-deep-convolutional","title":"A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection","date":"2016-07-25","arxiv_id":"1607.07155","repositories_listed":1,"syntology":null}],"syntology_records":10,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}